Re: [PATCH v14 for 4.1] sys_membarrier(): system-wide memory barrier (x86)

[Paul E. McKenney]

On Wed, Mar 25, 2015 at 03:03:47PM -0400, Mathieu Desnoyers wrote:
> Here is an implementation of a new system call, sys_membarrier(), which
> executes a memory barrier on all threads running on the system. It is
> implemented by calling synchronize_sched(). It can be used to distribute
> the cost of user-space memory barriers asymmetrically by transforming
> pairs of memory barriers into pairs consisting of sys_membarrier() and a
> compiler barrier. For synchronization primitives that distinguish
> between read-side and write-side (e.g. userspace RCU [1], rwlocks), the
> read-side can be accelerated significantly by moving the bulk of the
> memory barrier overhead to the write-side.

[ . . . ]

> Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@xxxxxxxxxxxx>
> CC: Paul E. McKenney <paulmck@xxxxxxxxxxxxxxxxxx>
> CC: Josh Triplett <josh@xxxxxxxxxxxxxxxx>
> CC: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
> CC: Steven Rostedt <rostedt@xxxxxxxxxxx>
> CC: Nicholas Miell <nmiell@xxxxxxxxxxx>
> CC: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
> CC: Ingo Molnar <mingo@xxxxxxxxxx>
> CC: Alan Cox <gnomes@xxxxxxxxxxxxxxxxxxx>
> CC: Lai Jiangshan <laijs@xxxxxxxxxxxxxx>
> CC: Stephen Hemminger <stephen@xxxxxxxxxxxxxxxxxx>
> CC: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
> CC: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> CC: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> CC: David Howells <dhowells@xxxxxxxxxx>

Reviewed-by: Paul E. McKenney <paulmck@xxxxxxxxxxxxxxxxxx>

> ---
>  MAINTAINERS                       |    8 ++++
>  arch/x86/syscalls/syscall_32.tbl  |    1 +
>  arch/x86/syscalls/syscall_64.tbl  |    1 +
>  include/linux/syscalls.h          |    2 +
>  include/uapi/asm-generic/unistd.h |    4 +-
>  include/uapi/linux/Kbuild         |    1 +
>  include/uapi/linux/membarrier.h   |   57 ++++++++++++++++++++++++++++
>  init/Kconfig                      |   12 ++++++
>  kernel/Makefile                   |    1 +
>  kernel/membarrier.c               |   75 +++++++++++++++++++++++++++++++++++++
>  kernel/sys_ni.c                   |    3 +
>  11 files changed, 164 insertions(+), 1 deletions(-)
>  create mode 100644 include/uapi/linux/membarrier.h
>  create mode 100644 kernel/membarrier.c
> 
> diff --git a/MAINTAINERS b/MAINTAINERS
> index d66a97d..7fbb698 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -6206,6 +6206,14 @@ W:	http://www.mellanox.com
>  Q:	http://patchwork.ozlabs.org/project/netdev/list/
>  F:	drivers/net/ethernet/mellanox/mlx4/en_*
> 
> +MEMBARRIER SUPPORT
> +M:	Mathieu Desnoyers <mathieu.desnoyers@xxxxxxxxxxxx>
> +M:	"Paul E. McKenney" <paulmck@xxxxxxxxxxxxxxxxxx>
> +L:	linux-kernel@xxxxxxxxxxxxxxx
> +S:	Supported
> +F:	kernel/membarrier.c
> +F:	include/uapi/linux/membarrier.h
> +
>  MEMORY MANAGEMENT
>  L:	linux-mm@xxxxxxxxx
>  W:	http://www.linux-mm.org
> diff --git a/arch/x86/syscalls/syscall_32.tbl b/arch/x86/syscalls/syscall_32.tbl
> index b3560ec..439415f 100644
> --- a/arch/x86/syscalls/syscall_32.tbl
> +++ b/arch/x86/syscalls/syscall_32.tbl
> @@ -365,3 +365,4 @@
>  356	i386	memfd_create		sys_memfd_create
>  357	i386	bpf			sys_bpf
>  358	i386	execveat		sys_execveat			stub32_execveat
> +359	i386	membarrier		sys_membarrier
> diff --git a/arch/x86/syscalls/syscall_64.tbl b/arch/x86/syscalls/syscall_64.tbl
> index 8d656fb..823130d 100644
> --- a/arch/x86/syscalls/syscall_64.tbl
> +++ b/arch/x86/syscalls/syscall_64.tbl
> @@ -329,6 +329,7 @@
>  320	common	kexec_file_load		sys_kexec_file_load
>  321	common	bpf			sys_bpf
>  322	64	execveat		stub_execveat
> +323	common	membarrier		sys_membarrier
> 
>  #
>  # x32-specific system call numbers start at 512 to avoid cache impact
> diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
> index 85893d7..058ec0a 100644
> --- a/include/linux/syscalls.h
> +++ b/include/linux/syscalls.h
> @@ -882,4 +882,6 @@ asmlinkage long sys_execveat(int dfd, const char __user *filename,
>  			const char __user *const __user *argv,
>  			const char __user *const __user *envp, int flags);
> 
> +asmlinkage long sys_membarrier(int flags);
> +
>  #endif
> diff --git a/include/uapi/asm-generic/unistd.h b/include/uapi/asm-generic/unistd.h
> index e016bd9..8da542a 100644
> --- a/include/uapi/asm-generic/unistd.h
> +++ b/include/uapi/asm-generic/unistd.h
> @@ -709,9 +709,11 @@ __SYSCALL(__NR_memfd_create, sys_memfd_create)
>  __SYSCALL(__NR_bpf, sys_bpf)
>  #define __NR_execveat 281
>  __SC_COMP(__NR_execveat, sys_execveat, compat_sys_execveat)
> +#define __NR_membarrier 282
> +__SYSCALL(__NR_membarrier, sys_membarrier)
> 
>  #undef __NR_syscalls
> -#define __NR_syscalls 282
> +#define __NR_syscalls 283
> 
>  /*
>   * All syscalls below here should go away really,
> diff --git a/include/uapi/linux/Kbuild b/include/uapi/linux/Kbuild
> index 00b10002..c5b0dbf 100644
> --- a/include/uapi/linux/Kbuild
> +++ b/include/uapi/linux/Kbuild
> @@ -248,6 +248,7 @@ header-y += mdio.h
>  header-y += media.h
>  header-y += media-bus-format.h
>  header-y += mei.h
> +header-y += membarrier.h
>  header-y += memfd.h
>  header-y += mempolicy.h
>  header-y += meye.h
> diff --git a/include/uapi/linux/membarrier.h b/include/uapi/linux/membarrier.h
> new file mode 100644
> index 0000000..b6f8f40
> --- /dev/null
> +++ b/include/uapi/linux/membarrier.h
> @@ -0,0 +1,57 @@
> +#ifndef _UAPI_LINUX_MEMBARRIER_H
> +#define _UAPI_LINUX_MEMBARRIER_H
> +
> +/*
> + * linux/membarrier.h
> + *
> + * membarrier system call API
> + *
> + * Copyright (c) 2010, 2015 Mathieu Desnoyers <mathieu.desnoyers@xxxxxxxxxxxx>
> + *
> + * Permission is hereby granted, free of charge, to any person obtaining a copy
> + * of this software and associated documentation files (the "Software"), to deal
> + * in the Software without restriction, including without limitation the rights
> + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
> + * copies of the Software, and to permit persons to whom the Software is
> + * furnished to do so, subject to the following conditions:
> + *
> + * The above copyright notice and this permission notice shall be included in
> + * all copies or substantial portions of the Software.
> + *
> + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
> + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
> + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
> + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
> + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
> + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
> + * SOFTWARE.
> + */
> +
> +/*
> + * All memory accesses performed in program order from each thread on
> + * the system is guaranteed to be ordered with respect to sys_membarrier().
> + * If we use the semantic "barrier()" to represent a compiler barrier
> + * forcing memory accesses to be performed in program order across the
> + * barrier, and smp_mb() to represent explicit memory barriers forcing
> + * full memory ordering across the barrier, we have the following
> + * ordering table for each pair of barrier(), sys_membarrier() and
> + * smp_mb() :
> + *
> + * The pair ordering is detailed as (O: ordered, X: not ordered):
> + *
> + *                        barrier()   smp_mb() sys_membarrier()
> + *        barrier()          X           X            O
> + *        smp_mb()           X           O            O
> + *        sys_membarrier()   O           O            O
> + */
> +
> +/* System call membarrier "flags" argument. */
> +enum {
> +	/*
> +	 * Query whether the rest of the specified flags are supported,
> +	 * without performing synchronization.
> +	 */
> +	MEMBARRIER_QUERY = (1 << 31),
> +};
> +
> +#endif /* _UAPI_LINUX_MEMBARRIER_H */
> diff --git a/init/Kconfig b/init/Kconfig
> index 9afb971..2452f3c 100644
> --- a/init/Kconfig
> +++ b/init/Kconfig
> @@ -1568,6 +1568,18 @@ config PCI_QUIRKS
>  	  bugs/quirks. Disable this only if your target machine is
>  	  unaffected by PCI quirks.
> 
> +config MEMBARRIER
> +	bool "Enable membarrier() system call" if EXPERT
> +	default y
> +	help
> +	  Enable the membarrier() system call that allows issuing memory
> +	  barriers across all running threads, which can be used to distribute
> +	  the cost of user-space memory barriers asymmetrically by transforming
> +	  pairs of memory barriers into pairs consisting of membarrier() and a
> +	  compiler barrier.
> +
> +	  If unsure, say Y.
> +
>  config EMBEDDED
>  	bool "Embedded system"
>  	option allnoconfig_y
> diff --git a/kernel/Makefile b/kernel/Makefile
> index a59481a..b572ced 100644
> --- a/kernel/Makefile
> +++ b/kernel/Makefile
> @@ -95,6 +95,7 @@ obj-$(CONFIG_CRASH_DUMP) += crash_dump.o
>  obj-$(CONFIG_JUMP_LABEL) += jump_label.o
>  obj-$(CONFIG_CONTEXT_TRACKING) += context_tracking.o
>  obj-$(CONFIG_TORTURE_TEST) += torture.o
> +obj-$(CONFIG_MEMBARRIER) += membarrier.o
> 
>  $(obj)/configs.o: $(obj)/config_data.h
> 
> diff --git a/kernel/membarrier.c b/kernel/membarrier.c
> new file mode 100644
> index 0000000..3077e94
> --- /dev/null
> +++ b/kernel/membarrier.c
> @@ -0,0 +1,75 @@
> +/*
> + * Copyright (C) 2010, 2015 Mathieu Desnoyers <mathieu.desnoyers@xxxxxxxxxxxx>
> + *
> + * membarrier system call
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License, or
> + * (at your option) any later version.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + */
> +
> +#include <linux/syscalls.h>
> +#include <linux/membarrier.h>
> +
> +static int membarrier_validate_flags(int flags)
> +{
> +	/* Check for unrecognized flags. */
> +	if (flags & ~MEMBARRIER_QUERY)
> +		return -EINVAL;
> +	return 0;
> +}
> +
> +#ifdef CONFIG_SMP
> +
> +/*
> + * sys_membarrier - issue memory barrier on all running threads
> + * @flags: MEMBARRIER_QUERY:
> + *             Query whether the rest of the specified flags are supported,
> + *             without performing synchronization.
> + *
> + * return values: Returns -EINVAL if the flags are incorrect. Testing
> + * for kernel sys_membarrier support can be done by checking for -ENOSYS
> + * return value.  Return value of 0 indicates success. For a given set
> + * of flags on a given kernel, this system call will always return the
> + * same value. It is therefore correct to check the return value only
> + * once during a process lifetime, setting MEMBARRIER_QUERY to only
> + * check if the flags are supported, without performing any
> + * synchronization.
> + *
> + * This system call executes a memory barrier on all running threads.
> + * Upon completion, the caller thread is ensured that all running
> + * threads have passed through a state where all memory accesses to
> + * user-space addresses match program order. (non-running threads are de
> + * facto in such a state.)
> + *
> + * On uniprocessor systems, this system call simply returns 0 after
> + * validating the arguments, so user-space knows it is implemented.
> + */
> +SYSCALL_DEFINE1(membarrier, int, flags)
> +{
> +	int retval;
> +
> +	retval = membarrier_validate_flags(flags);
> +	if (retval)
> +		goto end;
> +	if (unlikely(flags & MEMBARRIER_QUERY) || num_online_cpus() == 1)
> +		goto end;
> +	synchronize_sched();
> +end:
> +	return retval;
> +}
> +
> +#else /* !CONFIG_SMP */
> +
> +SYSCALL_DEFINE1(membarrier, int, flags)
> +{
> +	return membarrier_validate_flags(flags);
> +}
> +
> +#endif /* CONFIG_SMP */
> diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
> index 5adcb0a..5913b84 100644
> --- a/kernel/sys_ni.c
> +++ b/kernel/sys_ni.c
> @@ -229,3 +229,6 @@ cond_syscall(sys_bpf);
> 
>  /* execveat */
>  cond_syscall(sys_execveat);
> +
> +/* membarrier */
> +cond_syscall(sys_membarrier);
> -- 
> 1.7.7.3
> 

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/